Fifty percent of Medicaid patients treated on hospital psychiatric units fail to attend a follow-up outpatient service within 30 days of discharge. These failed care transitions increase the risk of relapse, readmission, criminal justice involvement, and suicide. Hospital care transition practices meant to facilitate transitions include communicating with outpatient mental health providers, scheduling timely post-discharge outpatient appointments, and forwarding treatment summaries to outpatient providers. These practices represent a standard of care and are included in many new public reporting and pay-for-performance programs. However, surprisingly little is known about how often hospital providers complete these practices, which practices are more critical than others, and for which patients these practices may not be sufficient to ensure follow-up. The proposed project addresses these questions by examining hospital care transition practices for over 30,000 Medicaid discharges from more than 100 hospital psychiatric units throughout New York State (NYS) in 2012 - 2013. As part of a statewide managed care readiness program, hospitals were required to notify a contracted managed behavioral health organization (MBHO) of Medicaid fee-for-service admissions to a hospital psychiatric unit and report whether they: 1) communicated with current or prior outpatient mental health providers; 2) scheduled a follow-up appointment with an outpatient provider; and 3) forwarded a case summary to the outpatient provider. The proposed project combines data from MBHOs with Medicaid claims data and information on hospital, outpatient provider, and regional service system characteristics. We will determine the rates at which hospital providers completed each of the three care transition practices and identify characteristics of patients that did and did not receive each practice. We will then determine the effectiveness of the care transition practices after controlling for patien characteristics (e.g., co-morbid general medical or substance use disorders, lack of contact with providers prior to admission), hospital characteristics (e.g., size, teaching status, presence of on-site outpatient services), characteristics of outpatient providers receiving referrals (e.g., sie, type, availability of care management supports), and regional service system characteristics (e.g., provider capacity, population density, per capita income) known to impact care transitions. We will determine whether the care transition practices were effective in high-need subgroups such as youth, patients with prior failure to attend outpatient care or frequent use of hospital services, and patients with co- morbid general medical disorders. Finally, we will identify patient subgroups for which the practices were insufficient and that are likely to require more intensive approaches to ensure successful care transitions. Findings from this project will guide provider efforts to improve care transitions, provide validity data for potential new quality measures related to mental health care transitions, and inform system level quality improvement initiatives that will further strengthen safety net services for Medicaid patients.